Computing Reviews

Typing Tutor:individualized tutoring in text entry for older adults based on input stumble detection
Hagiya T., Horiuchi T., Yazaki T.  CHI 2016 (Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems, Santa Clara, CA, May 7-12, 2016)733-744,2016.Type:Proceedings
Date Reviewed: 07/26/16

As older users adopt smartphones, they need to acquire new skills for interacting with touch-based devices. This paper describes the development of a support system, Typing Tutor, to improve text entry for older users in Japan transitioning from feature phones to smartphones. After confirming that older users are amenable to instructions (as opposed to trial and error, for example), and identifying instances and categories of “input stumbles,” the authors developed and evaluated the Typing Tutor system.

In contrast to mistyping by targeting one key but hitting another, input stumbles relate to selecting suggestions; using special keys like Enter, Delete, or modifier keys; or entering symbols. These input stumbles are somewhat language dependent, especially for languages that use multibyte encodings for characters.

Typing Tutor builds a user model that continually adjusts aspects such as the user’s skill level. This model is used to generate instructions for situations where a user encounters a problem and is likely to benefit from an instruction.

A final experiment compared the performance of two groups of users, one transitioning to a smartphone with Typing Tutor and the other without. The results confirm the effectiveness of the system in terms of typing proficiency measured by the number and types of input stumbles, while other aspects not addressed by Typing Tutor like accuracy or typing speed did not vary significantly.

Overall, individualized, targeted assistance with text entry issues looks like a promising building block to ease the transition to smartphones for older users. While there are some restrictions to the Typing Tutor model based on the language used, similar models can be developed for other languages.

Reviewer:  Franz Kurfess Review #: CR144631 (1611-0844)

Reproduction in whole or in part without permission is prohibited.   Copyright 2024 ComputingReviews.com™
Terms of Use
| Privacy Policy